Electrical Impedance Tomography (EIT) is a new medical imaging method which, uniquely, offers a way to image neuronal depolarization in nervous tissue using arrays of external electrodes. In rat cerebral cortex, it has a resolution of <200m and 1 ms using epicortical electrode arrays during sensory evoked potentials. It has been pioneered by PI David Holder at UCL London who has recently extended it to imaging compound action potential traffic within nerves with a flexible silicone rubber cylindrical cuff. The same cuff may then also be used for selective stimulation of the identified fascicles. This has been supported by Galvani Bioelectronics as it could furnish an essential way to avoid off-target effects in Electroceutical stimulation of autonomic nerve. Until now, it has been undertaken using an external cylindrical cuff with 16-32 electrodes. Spatial resolution is good at the edge of the nerve near the electrodes but decreases towards the centre of the nerve where current density is less. Our objective is to improve spatial resolution by adapting the cuff to have a ?cartwheel? design with spokes extending into the interior of the nerve. Penetrating intraneural electrodes are rigid and cause nerve damage, inflammation, and loss of recording ability. However, we mitigate this risk by the use of novel ?shape memory polymer? pioneered by co-PI Walter Voit at University of Texas Dallas (UTD). This is rigid at room temperature, so permitting insertion, but becomes flexible at body temperature so that it is biocompatible and does not elicit an inflammatory response. This would permit higher spatial resolution in the interior of the nerve in imaging while also offering a way to achieve selective stimulation of identified fascicles impossible with an external cylindrical cuff alone. In year 1 we will undertake preliminary studies with different electrode materials and a preliminary geometry of 4 spokes with 4 electrodes each. This will be tested in saline filled tanks, sheep vagus ex and in vivo, and rabbit sciatic nerve in vivo. In Year 2, further acute studies in rabbit sciatic and sheep vagus nerve during evoked activity will be undertaken with an improved electrode geometry suggested by modelling and empirical studies in Year 1, with up to 16 spokes with 16 electrodes each. In Year 3, we will evaluate the biocompatibility and performance in imaging in chronic studies over up to 3 months in sheep vagus nerve and an acute pilot study in human vagus nerve in subjects where the nerve is exposed for insertion of a vagal nerve stimulator. In parallel, we will optimise EIT imaging reconstruction methods, combine EIT academic software into a robust package ready for use in clinical trials, and develop a mechanical electrode insertion rig which can be inserted through an endoscope. In all animal studies, performance will be assessed by histology, electrophysiology and EIT imaging. The deliverable will be a novel ?cartwheel? electrode design for high resolution EIT imaging in nerve, and a rigorous evaluation of its performance and biocompatibility. We envisage that it will cause negligible nerve damage when used over months and deliver imaging of fascicular compound action potentials throughout the cross section of the nerve with an accuracy of <150m and 1ms for human vagus nerve.